PERCEPT: A New Online Change-Point Detection Method using Topological Data Analysis

PERCEPT: A New Online Change-Point Detection Method using Topological Data Analysis
复制标题

PERCEPT:一种利用拓扑数据分析的新型在线变点检测方法

DOI:
10.1080/00401706.2022.2124312
复制
发表时间:
2023
期刊:
影响因子:
2.5
通讯作者:
Xie, Yao
Xie, Yao
中科院分区:
工程技术3区
文献类型:
--
作者:
Zheng, Xiaojun;Mak, Simon;Xie, Liyan;Xie, Yao

文献摘要

参考文献

被引文献

相似文献

拓扑数据分析(TDA)提供了一套数据分析工具,用于从复杂的高维数据集中提取嵌入的拓扑结构。近年来,TDA已经成为一个快速发展的领域,在广泛的应用中取得了成功,包括信号处理,神经科学和网络分析。在这些应用中,变化的在线检测是至关重要的,但这可能是非常具有挑战性的,因为这种变化往往发生在高维数据流中的低维嵌入。因此,我们提出了一种新的方法,称为基于持久性的变化点检测(PERCEPT),它利用从TDA学习的拓扑结构,顺序检测变化。PERCEPT遵循两个关键步骤:它首先通过持久性图学习作为点云的嵌入式拓扑,然后应用非参数监测方法来检测所得到的点云分布的变化。这产生了一个非参数的,拓扑感知的框架,可以有效地检测在线几何变化。我们调查的有效性的感知现有的方法在一套数值实验中的数据流有一个嵌入式的拓扑结构。然后,我们证明了在太阳耀斑监测和人体姿态检测的两个应用程序中的有用性。
Topological data analysis (TDA) provides a set of data analysis tools for extracting embedded topological structures from complex high-dimensional datasets. In recent years, TDA has been a rapidly growing field which has found success in a wide range of applications, including signal processing, neuroscience and network analysis. In these applications, the online detection of changes is of crucial importance, but this can be highly challenging since such changes often occur in low-dimensional embeddings within high-dimensional data streams. We thus propose a new method, called PERsistence diagram-based ChangE-PoinT detection (PERCEPT), which leverages the learned topological structure from TDA to sequentially detect changes. PERCEPT follows two key steps: it first learns the embedded topology as a point cloud via persistence diagrams, then applies a nonparametric monitoring approach for detecting changes in the resulting point cloud distributions. This yields a nonparametric, topology-aware framework which can efficiently detect online geometric changes. We investigate the effectiveness of PERCEPT over existing methods in a suite of numerical experiments where the data streams have an embedded topological structure. We then demonstrate the usefulness of PERCEPT in two applications on solar flare monitoring and human gesture detection.
DOI: 10.1080/01621459.2017.1409123
发表时间: 2016-11
影响因子: 3.7
作者:
Simon Mak;Chih-Li Sung;Xingjian Wang;Shiang-Ting Yeh;Yu-Hung Chang;V. R. Joseph;V. Yang;C. F. J. Wu
通讯作者: Simon Mak;Chih-Li Sung;Xingjian Wang;Shiang-Ting Yeh;Yu-Hung Chang;V. R. Joseph;V. Yang;C. F. J. Wu
DOI: 10.21105/joss.00925
发表时间: 2018-09
期刊: J. Open Source Softw.
影响因子: --
作者:
Christopher J. Tralie;Nathaniel Saul;R. Bar-On
通讯作者: Christopher J. Tralie;Nathaniel Saul;R. Bar-On
DOI: 10.1201/b17279
发表时间: 2014-08
期刊: Food Biophysics
影响因子: 3
作者:
A. Tartakovsky;I. Nikiforov;M. Basseville
通讯作者: A. Tartakovsky;I. Nikiforov;M. Basseville
滑动窗口和持久性
DOI: 10.1121/1.4987655
发表时间: 2017
影响因子: 2.4
作者:
Jose A. Perea;Chris Traile
通讯作者: Chris Traile
DOI: 10.1007/s10208-014-9206-z
发表时间: 2015-06-01
影响因子: 3
作者:
Perea, Jose A.;Harer, John
通讯作者: Harer, John